 So I've been incredibly impressed with the support that this program has been given over the last eight years or so, and really that's, there's testament to that and the people around this room. They're really an impressive group of people. I really want to thank you for coming. First one, right, thanks. So, and the other thing is, I guess it's good that Eric didn't say that I was the scarecrow in the Wizard of Oz because then I wished I'd had a brain and today I'm going to tell you about how the UDP worked and how the UDN is now working and how we can map the UDP onto the UDN or should we essentially and I would say that this is a non-surjective function. So how many people in the room would know what a non-surjective function is? Good, because I didn't either. Tom Markello told me what it is. It's a function that doesn't map every X to a Y. So it's non-surjective. So we don't really think of those functions very often. So let me go through how we established the UDP and what it's characterized by and then how the UDN is moving from that or towards that. The UDP was established to help patients reach a diagnosis and also discover new diseases and because we're the NIH, we have to do some research and some science. So the way it works is that people apply and they apply with their doctor's recommendation, a letter. And then for the adults, I look at the records and for their kids, Cindy Tiff looks at the records, we give them the consultants around the NIH who are all intramural and then they tell us what they think, you know, it's just a suitable case or isn't it? And we reject about 75, 80% of the patients. We actually accept about 50% of the kids and only about 20% of the adults and you can sort of figure out why that might be in terms of the balance of subjective complaints versus objective findings. And then we give some advice to some of the patients, but we see patients for one week as inpatients at the clinical center and it doesn't cost them anything. We even pay for their travel. And then we do incredible phenotyping because the NIH clinical center has an infrastructure that allows that and genetic studies which can be commercial. We do SNPs to essentially inform our exome analysis and then we do functional studies and some of them that we can. And the conclusions that we've drawn over the years are that phenotyping is critical and important and everybody knows that. So really the issue then is how can we achieve the best phenotyping? Half of our diagnosis are not made on next generation sequencing. They're made because we have doctors who come together or talk with each other, see the patients in an intense fashion for a week. Functional studies are limiting. They always have been and they always will because sometimes a functional study to determine whether a variant is actually pathogenic requires a year or two or three years of work by a postdoctoral fellow. Sharing of sequence and phenotyping information is essential and undiagnosed disease patients are desperate. So how does this convert now to the UDN? Strong phenotyping is critical. Again, everybody knows that. We do inpatient admissions at the NIH because we can and because we think that it's incredibly important for the patients to be available throughout an entire week for our consultants to come and meet other consultants in the hall and have meetings over these individuals. This is not something that may necessarily be convertible to an extramural system and we're finding that out. So I think that one of the things that we need to discuss here is for the sustainability of a program like this, how can we, let's say, achieve the essence of this program, basically having physicians talk together about a patient in an extramural setting? Nurse practitioners play a huge role in our system, either NPs or PAs. And this is not something that's necessarily done in every one of the clinical sites that's being supported by the UDN. I'll give you one example of why we think it's important. When we do phenotypes, which is the ontology system that needs to be entered for all of our patients for searching and comparability of cases, the people who fill out those, that phenotypes is largely nurse practitioners at the NIH. In other centers, if this is given to attending physicians, it takes longer to get done or sometimes it doesn't get done. I mean, even at our center, it doesn't get done well by the attending physicians. It gets done by the people who are seeing the patients for probably maybe 15 to 20 face hours in a week. Those are the nurse practitioners. Critically important element, which again may not be recapitulated extramirally. A friendly user ontology is essential and in fact the UDN has picked up the ontology that we picked up at the UDP. It's not necessarily 100% sure that that's the best, but the point is that an ontology system has been picked up that's been made universal. The second conclusion was half of our diagnoses are not based on next generation sequencing. There's a misconception sometimes that the UDP and the UDN is a sequencing program. It's not a sequencing program. The essence of this is that we see patients as physicians and we get other physicians to come together and look at the patients and make a diagnosis. It's also not solely a genetics program. We'll see patients who have any sort of illness as long as it's undiagnosed. And in fact, the UDN understands that and has an environmental survey involved with it, also model organisms and gene function cores. And again, it's not necessarily, it's solely a sequencing program, although we now have exome and genome sequencing cores that provide us with service in the Undiagnosed Disease Network. There are some sites that do sequence the patients before they see them. And we're gonna try to determine over the course of these phases, whether or not that's a beneficial thing or if it's not beneficial. We really don't know that yet. Functional studies are rate limiting and the Undiagnosed Disease Program has used many different ways to achieve this. For one thing, we have some in-house money and therefore we have in-house postdocs and postback students who work on some of the projects that emanate from our program. For another, we have contracts that we can let and also the UDN has let supplements to RO1s and also RFAs by the R21 mechanism and we have collaborators who work on projects for no money at all. And the collaborators are experts in the field and basically what they wanna do, I think, is to tie their research to a human disease and a human being for the basically introduction to their next RO1. The UDN itself has a gene function core and it has a model systems core and the clinical sites have the purview to use investigators at their own institutions to pursue the basic research into the variants that they find in the patients. Sharing of data is essential. So the UDP has shared its phenotypes and its variants in a variety of ways, on a national level and on an international level now. There is an undiagnosed disease network international that's been established and there's a white paper and molecular genetics and metabolism on that. We share through the coordinating center and the network shares through the coordinating center and can share internationally and this was founded in the institutional review board protocol that we wrote. In other words, the protocol and the consent allow for two things. One is the sharing of personally identified information among the members of the network. So that's a bit paradigm changing and it was approved by the NHGRI Institutional Review Board and it really is based on our experience that the patients wanna tell their stories. They wanna find another individual who has the same disorder, another family. And the second element is that the protocol and the consent allow for the sharing of de-identified information more broadly, that is with collaborators at the research level, not really necessarily the clinical level, but at the research level. And the UDN has used language basically drafted from the protocol that we wrote a long time ago and has actually changed over the years. And then the protocol then has the UDN has a single protocol that relies upon reliance agreements by every one of the sites that's involved in the UDN. And again, we share the information there. And finally, undiagnosed patients are desperate and this is a function that does map from the UDP to the UDN. Everybody recognizes this as soon as you see one of these individual patients. So that's what I've got. I think I made it short. In case there were questions, are you taking questions? Yeah, we have a few minutes if folks have some questions. Yes, sir. I apologize if everybody else knows this except me, but I'm not in the network. But is the consent form written broadly enough so that you don't have to ever consider going back for further permission? I mean, it's written broadly for application of all the data out to other applications. Yes, we think it is. I think, you know, since I wrote it, maybe that's a little bit of a biased view. But what do you think, Rachel? Do you think this is broadly enough written for future changes? It was intentionally written quite broadly, although there, of course, may be some circumstances in which we need to go back to patients, but intentionally for educational purposes, for sharing with outside databases, for sharing with other researchers, all of those are encompassed in the consent form. Give an example, in a way, we had to change our protocol about four years ago or so when IPS cells came in, and we incorporated that into the protocol and the consent. That's already in the UDN consent, but if something like that came along in the future, we may have to go and change it. Well, one thing I would ask specifically, is there a continuing interaction with those patients? I mean, can you incorporate some outcomes data or continuing data about the patients? Is there a provision for that? We're allowed to recontact them, yes. There now are patients in terms of us actually being doctors to them. I mean, they contact us all the time, and they can, and we can contact them. In the UDN as well, yes, that's correct. Hi, Judy. Do you have a patient family user group that sort of continually kind of asks the question of, or gives advice, or you can raise issues with? Well, right now we have Matt Might, who helps us with some of the social media and gives us a perspective from that standpoint. And written in is to establish such a group. I think I should ask Rachel how we're doing on that. So, yes, we're in the process of forming a patient engagement group specifically for that purpose. The question that is giving us a bit of pause is how to engage these families without overburdening them because we would be asking for some of their time. So, of course, you wouldn't have everybody, but it seems like some kind of user group that meets a couple of times a year to sort of say, okay, how things have gone. And part of it is that's become the way to do business. But when you set things up, it wasn't necessarily. But I think my guess is that the parent support groups would really love to see you do that. And the other part is newbies, like just the desperate new ones, have a different perspective than somebody who the diagnosis has been found, from somebody who the diagnosis didn't really help. But, yeah, I mean, I think they have very different perspectives over time. Agreed. Walter? Yeah, so maybe from the neurologic perspective, there are numerous patients that nobody knows what they have. They go around from doctor to doctor and everybody has the same questions. Everybody does the same kind of exam. And so the question comes up is, is there an effective way in which you could capture that information, maybe even in a video of the patient interview and have that kind of storage so that people could access it. And I mean, I'm not sure if it's helpful for the non-neurologic cases, but to be able to see that kind of interaction, get that history, I mean, you can almost think of putting it out and having people around the world try and take a shot at what the problem is. Is that kind of thing, do you think feasible? Well, so Dennis Landis, when he was actively with us, actually wrote a bunch of best practices for the investigation of an undiagnosed neurological case. That was written, it wasn't video. Are you sort of getting at doing this for an individual case where you'd like that case to be solved? Or are you asking more for general principles that are illustrated by a video of how to do an exam and what questions to ask for a new disease? You know, I'm thinking for an individual case because I'm just imagining if it may be off, but someone might see something and they have an idea, but it doesn't gel until they see the other one. So to be able to have a library of these kind of cases, because I think that's how people develop their expertise over time. They start seeing different things and they're kind of correlating. I saw this case before and this one looks like this and that's how things come together. So I'm just wondering in a collaborative way, thinking with modern techniques as opposed to sending the patient around to different doctors, you bring all the data and then different doctors can all look at the same data. Right, I think that that is in a way data sharing. And the way that we're sharing data now is to have the phenotypes, let's say categorized by an ontology that allows searchability and also to have the variants put in a database that's shared. What you're suggesting is going one step further, I think, at having videos and we do take videos of many of our patients, especially the very special ones. Disseminating those videos maybe is a little bit more difficult. In other words, videos by and large are not searchable. So in that case, we'd have to be a little bit selective about whom we would give the videos to for a particular consultation. But I absolutely agree that that is the most poignant way of finding similarities and that's critical. Do you want to add something, Rachel, before I go to? So at the Coordinating Center and Anastasia knows this, I don't know if you, Bill, yet are aware of this, but we have been exploring potential partnerships with networks of doctors that are doing collaborative case-solving to potentially post either solved UDN cases to educate a broader audience about undiagnosed diseases and how they should be worked up, as well as unsolved cases to crowdsource. And at the Department of Biomedical Informatics, where I was previously at Harvard, they had done crowdsourcing around undiagnosed diseases that was focused at sequencing cores. This would be more focused at clinicians. And so I think that that is a promising avenue and including the videos is very important. Matt Merritt. I was just gonna comment that in the VA system, we use telemedicine pretty heavily and also, so that could be a one-on-one with a patient at a distant site with their provider, also perhaps in the room. And it might be a way of actually not having the patient have to travel, but that they can get expertise from a whole bunch of other clinicians. You could even do it all at one time where you have multiple providers in different places with telehealth technology. And then there's also store and forward technology so that you could send images or videos for a consultation with another provider. So it's clinical care that is provided in certain health systems that have that capability. So just maybe something, I don't know if that's anything anyone has considered. So we're right at time now, but I know, John Mink, you've had your hand up. So why don't we take that as the last question for this session? Just very quickly, because I know this is a topic for discussion later today, but one of the things that makes undiagnosed patients desperate is that their physicians say, I don't know what you have, so I don't know how to treat it. And particularly for the UDP, the intramural program, what is typically done in terms of treatment recommendations once a diagnosis is made or are those recommendations even made? We do make recommendations largely asking that the local physician pick it up. In fact, that's one of the criteria for acceptance that there has to be someone who's willing to do that. But sometimes the disease is so new that there's no authority in it. And then we do take over some of that. And for example, in the last fiscal year, we saw 20 to 25 patients a second time, basically for that follow-up because there was some treatment that, and I would also mention that sometimes the ancillary services of the NIH Clinical Center are incredibly helpful in that. Rehabilitation medicine, for example, or some of the consultations in heme or immunology, things of that sort are extremely beneficial. Rheumatology, when they have new drugs and sometimes even the clinical center can provide those new drugs and the insurance won't. Just, although don't tell anybody at the clinical center that. But that's what we do. Unfortunately, with these genetic diseases, there may not be an awful lot of cases in which the therapy is directed. Thank you.